InforLorV4, PascalFrancis, Corpus, bibRecord, 000C34

High performance unconstrained word recognition system combining HMMs and Markov random fields

Identifieur interne : 000C34 ( PascalFrancis/Corpus ); précédent : 000C33; suivant : 000C35

High performance unconstrained word recognition system combining HMMs and Markov random fields

Auteurs : G. Saon ; A. Belaïd

Source :

International journal of pattern recognition and artificial intelligence [ 0218-0014 ] ; 1997.

RBID : Pascal:97-0491456

Descripteurs français

Pascal (Inist)
- Reconnaissance forme, Ecriture, Mot, Modèle Markov, Champ aléatoire, Chèque postal, Modèle Markov variable cachée.

English descriptors

KwdEn :
- Hand writing, Hidden Markov model, Markov model, Pattern recognition, Random field, Word.

Abstract

In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).

Notice en format standard (ISO 2709)

Pour connaître la documentation sur le format Inist Standard.

A01	`01`	`1`		`@0 0218-0014`
A03		`1`		`@0 Int. j. pattern recogn. artif. intell.`
A05				`@2 11`
A06				`@2 5`
A08	`01`	`1`	`ENG`	`@1 High performance unconstrained word recognition system combining HMMs and Markov random fields`
A09	`01`	`1`	`ENG`	`@1 Automatic bankcheck processing. Part II`
A11	`01`	`1`		`@1 SAON (G.)`
A11	`02`	`1`		`@1 BELAÏD (A.)`
A12	`01`	`1`		`@1 IMPEDOVO (Sebastiano) @9 ed.`
A12	`02`	`1`		`@1 WANG (P. S. P.) @9 ed.`
A12	`03`	`1`		`@1 BUNKE (H.) @9 ed.`
A14	`01`			`@1 CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239 @2 54506 Vandœuvre-Lès-Nancy @3 FRA @Z 1 aut. @Z 2 aut.`
A15	`01`			`@1 Dipartimento di Informatica, Università di Bari, Via Amendola 173 @2 70126 Bari @3 ITA @Z 1 aut.`
A15	`02`			`@1 Institut für Informatik und Angewandte Mathematik, Universität Bern, Neubrückstrasse 10 @2 3012 Bern @3 CHE @Z 3 aut.`
A15	`03`			`@1 College of Computer Science, Northeastern University, 360 Huntington Avenue @2 Boston, MA 02115 @3 USA @Z 2 aut.`
A20				`@1 771-788`
A21				`@1 1997`
A23	`01`			`@0 ENG`
A43	`01`			`@1 INIST @2 22088 @5 354000069329000050`
A44				`@0 0000 @1 © 1997 INIST-CNRS. All rights reserved.`
A45				`@0 26 ref.`
A47	`01`	`1`		`@0 97-0491456`
A60				`@1 P`
A61				`@0 A`
A64	`01`	`1`		`@0 International journal of pattern recognition and artificial intelligence`
A66	`01`			`@0 SGP`
C01	`01`		`ENG`	@0 In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).
C02	`01`	`X`		`@0 001D02C03`
C03	`01`	`1`	`FRE`	`@0 Reconnaissance forme @3 P @5 01`
C03	`01`	`1`	`ENG`	`@0 Pattern recognition @3 P @5 01`
C03	`02`	`X`	`FRE`	`@0 Ecriture @5 02`
C03	`02`	`X`	`ENG`	`@0 Hand writing @5 02`
C03	`02`	`X`	`SPA`	`@0 Escritura manual @5 02`
C03	`03`	`X`	`FRE`	`@0 Mot @5 03`
C03	`03`	`X`	`ENG`	`@0 Word @5 03`
C03	`03`	`X`	`SPA`	`@0 Palabra @5 03`
C03	`04`	`X`	`FRE`	`@0 Modèle Markov @5 04`
C03	`04`	`X`	`ENG`	`@0 Markov model @5 04`
C03	`04`	`X`	`SPA`	`@0 Modelo Markov @5 04`
C03	`05`	`X`	`FRE`	`@0 Champ aléatoire @5 05`
C03	`05`	`X`	`ENG`	`@0 Random field @5 05`
C03	`05`	`X`	`SPA`	`@0 Campo aleatorio @5 05`
C03	`06`	`1`	`FRE`	`@0 Chèque postal @4 INC @5 72`
C03	`07`	`1`	`FRE`	`@0 Modèle Markov variable cachée @4 CD @5 96`
C03	`07`	`1`	`ENG`	`@0 Hidden Markov model @4 CD @5 96`
N21				`@1 300`

Format Inist (serveur)

NO :	PASCAL 97-0491456 INIST
ET :	High performance unconstrained word recognition system combining HMMs and Markov random fields
AU :	SAON (G.); BELAÏD (A.); IMPEDOVO (Sebastiano); WANG (P. S. P.); BUNKE (H.)
AF :	CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239/54506 Vandœuvre-Lès-Nancy/France (1 aut., 2 aut.); Dipartimento di Informatica, Università di Bari, Via Amendola 173/70126 Bari/Italie (1 aut.); Institut für Informatik und Angewandte Mathematik, Universität Bern, Neubrückstrasse 10/3012 Bern/Suisse (3 aut.); College of Computer Science, Northeastern University, 360 Huntington Avenue/Boston, MA 02115/Etats-Unis (2 aut.)
DT :	Publication en série; Niveau analytique
SO :	International journal of pattern recognition and artificial intelligence; ISSN 0218-0014; Singapour; Da. 1997; Vol. 11; No. 5; Pp. 771-788; Bibl. 26 ref.
LA :	Anglais
EA :	In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).
CC :	001D02C03
FD :	Reconnaissance forme; Ecriture; Mot; Modèle Markov; Champ aléatoire; Chèque postal; Modèle Markov variable cachée
ED :	Pattern recognition; Hand writing; Word; Markov model; Random field; Hidden Markov model
SD :	Escritura manual; Palabra; Modelo Markov; Campo aleatorio
LO :	INIST-22088.354000069329000050
ID :	97-0491456

Links to Exploration step

Pascal:97-0491456

Le document en format XML

<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">High performance unconstrained word recognition system combining HMMs and Markov random fields</title>
<author><name sortKey="Saon, G" sort="Saon, G" uniqKey="Saon G" first="G." last="Saon">G. Saon</name>
<affiliation><inist:fA14 i1="01"><s1>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Belaid, A" sort="Belaid, A" uniqKey="Belaid A" first="A." last="Belaïd">A. Belaïd</name>
<affiliation><inist:fA14 i1="01"><s1>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">97-0491456</idno>
<date when="1997">1997</date>
<idno type="stanalyst">PASCAL 97-0491456 INIST</idno>
<idno type="RBID">Pascal:97-0491456</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000C34</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">High performance unconstrained word recognition system combining HMMs and Markov random fields</title>
<author><name sortKey="Saon, G" sort="Saon, G" uniqKey="Saon G" first="G." last="Saon">G. Saon</name>
<affiliation><inist:fA14 i1="01"><s1>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Belaid, A" sort="Belaid, A" uniqKey="Belaid A" first="A." last="Belaïd">A. Belaïd</name>
<affiliation><inist:fA14 i1="01"><s1>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">International journal of pattern recognition and artificial intelligence</title>
<title level="j" type="abbreviated">Int. j. pattern recogn. artif. intell.</title>
<idno type="ISSN">0218-0014</idno>
<imprint><date when="1997">1997</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">International journal of pattern recognition and artificial intelligence</title>
<title level="j" type="abbreviated">Int. j. pattern recogn. artif. intell.</title>
<idno type="ISSN">0218-0014</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Hand writing</term>
<term>Hidden Markov model</term>
<term>Markov model</term>
<term>Pattern recognition</term>
<term>Random field</term>
<term>Word</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Reconnaissance forme</term>
<term>Ecriture</term>
<term>Mot</term>
<term>Modèle Markov</term>
<term>Champ aléatoire</term>
<term>Chèque postal</term>
<term>Modèle Markov variable cachée</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>0218-0014</s0>
</fA01>
<fA03 i2="1"><s0>Int. j. pattern recogn. artif. intell.</s0>
</fA03>
<fA05><s2>11</s2>
</fA05>
<fA06><s2>5</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>High performance unconstrained word recognition system combining HMMs and Markov random fields</s1>
</fA08>
<fA09 i1="01" i2="1" l="ENG"><s1>Automatic bankcheck processing. Part II</s1>
</fA09>
<fA11 i1="01" i2="1"><s1>SAON (G.)</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>BELAÏD (A.)</s1>
</fA11>
<fA12 i1="01" i2="1"><s1>IMPEDOVO (Sebastiano)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="02" i2="1"><s1>WANG (P. S. P.)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="03" i2="1"><s1>BUNKE (H.)</s1>
<s9>ed.</s9>
</fA12>
<fA14 i1="01"><s1>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</fA14>
<fA15 i1="01"><s1>Dipartimento di Informatica, Università di Bari, Via Amendola 173</s1>
<s2>70126 Bari</s2>
<s3>ITA</s3>
<sZ>1 aut.</sZ>
</fA15>
<fA15 i1="02"><s1>Institut für Informatik und Angewandte Mathematik, Universität Bern, Neubrückstrasse 10</s1>
<s2>3012 Bern</s2>
<s3>CHE</s3>
<sZ>3 aut.</sZ>
</fA15>
<fA15 i1="03"><s1>College of Computer Science, Northeastern University, 360 Huntington Avenue</s1>
<s2>Boston, MA 02115</s2>
<s3>USA</s3>
<sZ>2 aut.</sZ>
</fA15>
<fA20><s1>771-788</s1>
</fA20>
<fA21><s1>1997</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>22088</s2>
<s5>354000069329000050</s5>
</fA43>
<fA44><s0>0000</s0>
<s1>© 1997 INIST-CNRS. All rights reserved.</s1>
</fA44>
<fA45><s0>26 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>97-0491456</s0>
</fA47>
<fA60><s1>P</s1>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>International journal of pattern recognition and artificial intelligence</s0>
</fA64>
<fA66 i1="01"><s0>SGP</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02C03</s0>
</fC02>
<fC03 i1="01" i2="1" l="FRE"><s0>Reconnaissance forme</s0>
<s3>P</s3>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="1" l="ENG"><s0>Pattern recognition</s0>
<s3>P</s3>
<s5>01</s5>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Ecriture</s0>
<s5>02</s5>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Hand writing</s0>
<s5>02</s5>
</fC03>
<fC03 i1="02" i2="X" l="SPA"><s0>Escritura manual</s0>
<s5>02</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Mot</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Word</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Palabra</s0>
<s5>03</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Modèle Markov</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Markov model</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Modelo Markov</s0>
<s5>04</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Champ aléatoire</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Random field</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Campo aleatorio</s0>
<s5>05</s5>
</fC03>
<fC03 i1="06" i2="1" l="FRE"><s0>Chèque postal</s0>
<s4>INC</s4>
<s5>72</s5>
</fC03>
<fC03 i1="07" i2="1" l="FRE"><s0>Modèle Markov variable cachée</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="07" i2="1" l="ENG"><s0>Hidden Markov model</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fN21><s1>300</s1>
</fN21>
</pA>
</standard>
<server><NO>PASCAL 97-0491456 INIST</NO>
<ET>High performance unconstrained word recognition system combining HMMs and Markov random fields</ET>
<AU>SAON (G.); BELAÏD (A.); IMPEDOVO (Sebastiano); WANG (P. S. P.); BUNKE (H.)</AU>
<AF>CRIN-CNRS, Bât Loria, Campus Scientifique, B.P. 239/54506 Vandœuvre-Lès-Nancy/France (1 aut., 2 aut.); Dipartimento di Informatica, Università di Bari, Via Amendola 173/70126 Bari/Italie (1 aut.); Institut für Informatik und Angewandte Mathematik, Universität Bern, Neubrückstrasse 10/3012 Bern/Suisse (3 aut.); College of Computer Science, Northeastern University, 360 Huntington Avenue/Boston, MA 02115/Etats-Unis (2 aut.)</AF>
<DT>Publication en série; Niveau analytique</DT>
<SO>International journal of pattern recognition and artificial intelligence; ISSN 0218-0014; Singapour; Da. 1997; Vol. 11; No. 5; Pp. 771-788; Bibl. 26 ref.</SO>
<LA>Anglais</LA>
<EA>In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP (Service de Recherche Technique de la Poste) French postal check database (7031 words, 1779 amounts, different scriptors).</EA>
<CC>001D02C03</CC>
<FD>Reconnaissance forme; Ecriture; Mot; Modèle Markov; Champ aléatoire; Chèque postal; Modèle Markov variable cachée</FD>
<ED>Pattern recognition; Hand writing; Word; Markov model; Random field; Hidden Markov model</ED>
<SD>Escritura manual; Palabra; Modelo Markov; Campo aleatorio</SD>
<LO>INIST-22088.354000069329000050</LO>
<ID>97-0491456</ID>
</server>
</inist>
</record>

Pour manipuler ce document sous Unix (Dilib)

EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/PascalFrancis/Corpus

HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000C34 | SxmlIndent | more

HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Corpus/biblio.hfd -nk 000C34 | SxmlIndent | more

Pour mettre un lien sur cette page dans le réseau Wicri

{{Explor lien
   |wiki=    Wicri/Lorraine
   |area=    InforLorV4
   |flux=    PascalFrancis
   |étape=   Corpus
   |type=    RBID
   |clé=     Pascal:97-0491456
   |texte=   High performance unconstrained word recognition system combining HMMs and Markov random fields
}}

This area was generated with Dilib version V0.6.33.
Data generation: Mon Jun 10 21:56:28 2019. Site generation: Fri Feb 25 15:29:27 2022

	Serveur d'exploration sur la recherche en informatique en Lorraine
	Attention, ce site est en cours de développement ! Attention, site généré par des moyens informatiques à partir de corpus bruts. Les informations ne sont donc pas validées.

Serveur d'exploration sur la recherche en informatique en Lorraine

High performance unconstrained word recognition system combining HMMs and Markov random fields

High performance unconstrained word recognition system combining HMMs and Markov random fields

Source :

Descripteurs français

English descriptors

Abstract

Notice en format standard (ISO 2709)

Format Inist (serveur)

Links to Exploration step

Le document en format XML

Pour manipuler ce document sous Unix (Dilib)

Pour mettre un lien sur cette page dans le réseau Wicri